NLTK 05 - Synsets, Hypernyms, Hyponyms

This is the fifth article on the series on NLTK - Natural Language Toolkit. In the previous article, I wrote about Stop words that are useful to filter out useless words from a paragraph to extract the context or meaning.

The next step after you identify stop words and remove them, is identifying the words that pass the filter, those that are left out ready to be analyzed. First off, what you do is, connect your program with a lexicon. A lexicon is like a dictionary, it has a lot of word structures, meanings, relationships etc.

So a word can be a part of something called a synset, it is basically a specific word which carries specific meaning to itself, along with the part of speech it belongs to.

I further talk about hypernyms and hyponyms. Hypernyms are words that are more abstract than a specified word. Like the hypernym of a 'cat' can be 'animal'. Hyponym is something that falls in the reverse direction, something that is more specific than a given word.

These concepts are pretty good and they are better understood by watching the video:

Leave a comment below :)

Rocky DeRaze is a man on a mission, he wants to provide free education via the internet on various topics to everyone on the planet, not for philanthropic reasons, but for the rush that comes with teaching.